Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    verbs in the past tense ending with 'ed' or 'd'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter 'd'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    EStream
    -0.87
     Remem
    -0.78
     Pastebin
    -0.73
     Nun
    -0.71
    éĹĺ
    -0.64
     reusable
    -0.64
     momentum
    -0.62
     constitu
    -0.62
     RIS
    -0.62
     mosqu
    -0.61
    POSITIVE LOGITS
    ynamic
    1.18
    olph
    1.14
    rown
    1.13
    iamond
    1.11
    ownt
    1.11
    etermined
    1.11
    olf
    1.10
    hyde
    1.08
    rench
    1.03
    elta
    1.03
    Act Density 0.036%

    No Known Activations